Factors Influencing Difficulty of Laparoscopic Abdominoperineal Resection for Ultra-Low Rectal Cancer

Development and validation of machine learning models and nomograms for predicting the surgical difficulty of laparoscopic resection in rectal cancer

BACKGROUND

The objective of this study is to develop and validate a machine learning (ML) prediction model for the assessment of laparoscopic total mesorectal excision (LaTME) surgery difficulty, as well as to identify independent risk factors that influence surgical difficulty. Establishing a nomogram aims to assist clinical practitioners in formulating more effective surgical plans before the procedure.

METHODS

This study included 186 patients with rectal cancer who underwent LaTME from January 2018 to December 2020. They were divided into a training cohort (n = 131) versus a validation cohort (n = 55). The difficulty of LaTME was defined based on Escal's et al. scoring criteria with modifications. We utilized Lasso regression to screen the preoperative clinical characteristic variables and intraoperative information most relevant to surgical difficulty for the development and validation of four ML models: logistic regression (LR), support vector machine (SVM), random forest (RF), and decision tree (DT). The performance of the model was assessed based on the area under the receiver operating characteristic curve(AUC), sensitivity, specificity, and accuracy. Logistic regression-based column-line plots were created to visualize the predictive model. Consistency statistics (C-statistic) and calibration curves were used to discriminate and calibrate the nomogram, respectively.

RESULTS

In the validation cohort, all four ML models demonstrate good performance: SVM AUC = 0.987, RF AUC = 0.953, LR AUC = 0.950, and DT AUC = 0.904. To enhance visual evaluation, a logistic regression-based nomogram has been established. Predictive factors included in the nomogram are body mass index (BMI), distance between the tumor to the dentate line ≤ 10 cm, radiodensity of visceral adipose tissue (VAT), area of subcutaneous adipose tissue (SAT), tumor diameter >3 cm, and comorbid hypertension.

CONCLUSION

In this study, four ML models based on intraoperative and preoperative risk factors and a nomogram based on logistic regression may be of help to surgeons in evaluating the surgical difficulty before operation and adopting appropriate responses and surgical protocols.

Interpretable machine learning model to predict surgical difficulty in laparoscopic resection for rectal cancer

Background

Laparoscopic total mesorectal excision (LaTME) is standard surgical methods for rectal cancer, and LaTME operation is a challenging procedure. This study is intended to use machine learning to develop and validate prediction models for surgical difficulty of LaTME in patients with rectal cancer and compare these models' performance.

Methods

We retrospectively collected the preoperative clinical and MRI pelvimetry parameter of rectal cancer patients who underwent laparoscopic total mesorectal resection from 2017 to 2022. The difficulty of LaTME was defined according to the scoring criteria reported by Escal. Patients were randomly divided into training group (80%) and test group (20%). We selected independent influencing features using the least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression method. Adopt synthetic minority oversampling technique (SMOTE) to alleviate the class imbalance problem. Six machine learning model were developed: light gradient boosting machine (LGBM); categorical boosting (CatBoost); extreme gradient boost (XGBoost), logistic regression (LR); random forests (RF); multilayer perceptron (MLP). The area under receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity and F1 score were used to evaluate the performance of the model. The Shapley Additive Explanations (SHAP) analysis provided interpretation for the best machine learning model. Further decision curve analysis (DCA) was used to evaluate the clinical manifestations of the model.

Results

A total of 626 patients were included. LASSO regression analysis shows that tumor height, prognostic nutrition index (PNI), pelvic inlet, pelvic outlet, sacrococcygeal distance, mesorectal fat area and angle 5 (the angle between the apex of the sacral angle and the lower edge of the pubic bone) are the predictor variables of the machine learning model. In addition, the correlation heatmap shows that there is no significant correlation between these seven variables. When predicting the difficulty of LaTME surgery, the XGBoost model performed best among the six machine learning models (AUROC=0.855). Based on the decision curve analysis (DCA) results, the XGBoost model is also superior, and feature importance analysis shows that tumor height is the most important variable among the seven factors.

Conclusions

This study developed an XGBoost model to predict the difficulty of LaTME surgery. This model can help clinicians quickly and accurately predict the difficulty of surgery and adopt individualized surgical methods.

Development of artificial blood loss and duration of excision score to evaluate surgical difficulty of total laparoscopic anterior resection in rectal cancer

Purpose

Total laparoscopic anterior resection (tLAR) has been gradually applied in the treatment of rectal cancer (RC). This study aims to develop a scoring system to predict the surgical difficulty of tLAR.

Methods

RC patients treated with tLAR were collected. The blood loss and duration of excision (BLADE) scoring system was built to assess the surgical difficulty by using restricted cubic spline regression. Multivariate logistic regression was used to evaluate the effect of the BLADE score on postoperative complications. The random forest (RF) algorithm was used to establish a preoperative predictive model for the BLADE score.

Results

A total of 1,994 RC patients were randomly selected for the training set and the test set, and 325 RC patients were identified as the external validation set. The BLADE score, which was built based on the thresholds of blood loss (60 ml) and duration of surgical excision (165 min), was the most important risk factor for postoperative complications. The areas under the curve of the predictive RF model were 0.786 in the training set, 0.640 in the test set, and 0.665 in the external validation set.

Conclusion

This preoperative predictive model for the BLADE score presents clinical feasibility and reliability in identifying the candidates to receive tLAR and in making surgical plans for RC patients.

Increased body mass index is associated with operative difficulty during robot‐assisted radical prostatectomy

Objective

This study aimed to identify factors associated with surgeon perception of robot‐assisted radical prostatectomy (RARP) difficulty.

Patients and Methods

This study surveyed surgeons performing RARP between 2017 and 2018 and asked them to rate operative conditions and difficulty as optimal, good, acceptable, or poor. These answers were stratified as optimal or suboptimal for this study. Associations between surgeon responses and variables hypothesized to affect surgical difficulty, including anatomic factors such as pelvic diameter and prostate volume:pelvic diameter ratio, were assessed.

Results

Between November 2017 and September 2018, a total of 100 patients were prospectively enrolled in the study of which 58 cases were rated as optimal and 42 were rated as suboptimal. Of the evaluated variables, only increasing clinical T stage (odds ratio [OR] 1.49, 95% confidence interval [CI] 1.03–2.15, p = 0.03) and increasing body mass index (BMI) (OR 1.14, 95% CI 1.03–1.26, p = 0.01) were associated with increased difficulty; 90‐day complication rates were similar between the optimal and suboptimal cohorts (17.3% vs. 23.8%, respectively; p = 0.5). The number of patients with previous surgery, pelvic diameter, and prostate size:pelvic diameter ratio were not significantly different between cohorts (p > 0.05 for all). Operative time (ρ = 0.23, p = 0.02) and estimated blood loss (EBL) (ρ = 0.38, p = 0.0001) were correlated with suboptimal difficulty.

Conclusion

The factors associated with surgeon‐reported RARP difficulty were patient BMI and clinical T stage among surgeons with significant RARP experience. These data should be incorporated into surgical decision making and patient counseling prior to performing a RARP.

A nomogram predicting the difficulty of laparoscopic surgery for rectal cancer

PURPOSE

This study aimed to identify the risk factors associated with performing a difficult laparoscopic radical resection of rectal cancer, and to establish a predictive nomogram to help individual clinical treatment decisions.

METHODS

A total of 977 patients with rectal cancer who underwent laparoscopic radical resection between January 2014 and December 2016 were enrolled in this study. The difficulty of laparoscopic-assisted rectal resection (LARR) was defined according to the scoring criteria reported by Escal. A logistic regression analysis was performed to identify the variables that may affect the difficulty of LARR, and a nomogram predicting the surgical difficulty was created.

RESULTS

A multivariate analysis demonstrated that a BMI > 28 kg/m², the distance between the tumor and the anal margin ≤ 5 cm, the maximum transverse tumor diameter > 3 cm tumor, interspinous distance < 10 cm, history of abdominal surgery, and preoperative radiotherapy were independent risk factors and they were, therefore, included in the predictive nomogram for identifying a difficult LARR.

CONCLUSIONS

This study defined a difficult LARR and identified independent risk factors for a difficult operation and created a predictive nomogram for difficult LARR. This nomogram may facilitate the stratification of patients at risk for being associated with a difficult LARR for rectal cancer.

Trans-anal total mesorectal excision in low rectal cancers: Preliminary oncological results of a comparative study

OBJECTIVE

The management of lower rectal cancers is a therapeutic challenge both from the oncological and functional viewpoints. The aim of this study is to assess the oncological results and postoperative morbidity after transanal total mesorectal excision (TaTME) for low rectal cancer.

MATERIAL AND METHODS

In this monocentric retrospective study, we compared the quality of carcinologic resection and the morbidity-mortality between a group of 20 patients undergoing TaTME and 21 patients treated by abdomino-perineal resection (APR) between 2016 to 2019.

RESULTS

More patients had a positive circumferential resection margin (CRM) (≤1mm) in the APR group (47.6% vs. 5%; P<0.0036). The difference in the rates of grades I-II and III-IV complications (Clavien-Dindo classification) between the two groups was not statistically significant (50% vs. 57.1% and 5% vs. 9.5% in TaTME and APR, respectively; P=0.7579, P=1.00). The median follow-up was longer in the TaTME group (20 months vs. 11 months; P=0.58). The local recurrence rate did not differ between the two groups (5% vs. 4.8%; P=1.00) CONCLUSION: TaTME provides a reliable total mesorectal resection with an acceptable CRM. However, like any new technique, it requires experience and the learning curve is long.

Prediction of surgical difficulty in minimally invasive surgery for rectal cancer by use of MRI pelvimetry

Background

Technical difficulties in rectal surgery are often related to dissection in a limited surgical field. This study investigated the clinical value of MRI pelvimetry in the prediction of surgical difficulty associated with minimally invasive rectal surgery.

Methods

Patients with rectal cancer who underwent laparoscopic or robotic total mesorectal excision between 2005 and 2017 were reviewed retrospectively and categorized according to surgical difficulty on the basis of duration of surgery, conversion to an open procedure, use of the transanal approach, postoperative hospital stay, blood loss and postoperative complications. Preoperative clinical and MRI‐related parameters were examined to develop a prediction model to estimate the extent of surgical difficulty, and to compare anastomotic leakage rates in the low‐ and high‐grade surgical difficulty groups. Prognosis was investigated by calculating overall and relapse‐free survival, and cumulative local and distant recurrence rates.

Results

Of 121 patients analysed, 104 (86·0 per cent) were categorized into the low‐grade group and 17 (14·0 per cent) into the high‐grade group. Multivariable analysis indicated that high‐grade surgical difficulty was associated with a BMI above 25 kg/m² (odds ratio (OR) 4·45, P = 0·033), tumour size 45 mm or more (OR 5·42, P = 0·042), anorectal angle 123° or more (OR 5·98, P = 0·028) and pelvic outlet less than 82·7 mm (OR 6·62, P = 0·048). All of these features were used to devise a four‐variable scoring model to predict surgical difficulty. In patients categorized as high grade for surgical difficulty, the anastomotic leakage rate was 53 per cent (9 of 17 patients), compared with 9·6 per cent (10 of 104) in the low‐grade group (P < 0·001). The high‐grade group had a significantly higher local recurrence rate than the low‐grade group (P = 0·002).

Conclusion

This study highlights the impact of clinical variables and MRI pelvimetry in the prediction of surgical difficulty in minimally invasive rectal surgery.

Assessing surgical difficulty in locally advanced mid-low rectal cancer: the accuracy of two MRI-based predictive scores

AIM

Predicting surgical difficulty is a critical factor in the management of locally advanced rectal cancer (LARC). This study evaluates the accuracy and external validity of a recently published morphometric score to predict surgical difficulty and additionally proposes a new score to identify preoperatively LARC patients with a high risk of having a difficult surgery.

METHODS

This is a retrospective study based on the European MRI and Rectal Cancer Surgery (EuMaRCS) database, including patients with mid/low LARC who were treated with neoadjuvant chemoradiation therapy and laparoscopic total mesorectal excision (L-TME) with primary anastomosis. For all patients, pretreatment and restaging MRI were available. Surgical difficulty was graded as high and low based upon a composite outcome, including operative (e.g. duration of surgery) and postoperative variables (e.g. hospital stay). Score accuracy was assessed by estimating sensitivity, specificity and area under the receiver operating characteristic curve (AROC).

RESULTS

In a total of 136 LARC patients, 17 (12.5%) were graded as high surgical difficulty. The previously published score (calculated on body mass index, intertuberous distance, mesorectal fat area, type of anastomosis) showed low predictive value (sensitivity 11.8%; specificity 92.4%; AROC 0.612). The new EuMaRCS score was developed using the following significant predictors of surgical difficulty: body mass index > 30, interspinous distance < 96.4 mm, ymrT stage ≥ T3b and male sex. It demonstrated high accuracy (AROC 0.802).

CONCLUSION

The EuMaRCS score was found to be more sensitive and specific than the previous score in predicting surgical difficulty in LARC patients who are candidates for L-TME. However, this score has yet to be externally validated.